Children’s Emotional Well-Being, Difficulties, and Aggressive Behaviour during the First COVID-19 Pandemic Lockdown in German-Speaking Switzerland ()
1. Introduction
In spring 2020, many governments around the world took drastic measures to prevent the further spread of COVID-19. However, compared to other European countries, Switzerland introduced relatively lenient measures. For example, staying at home was recommended but not required; no curfews were imposed. Consequently, Swiss children were not restricted to their homes, as public spaces such as playgrounds remained open even during the nationwide lockdown. However, from mid-March to mid-May 2020 many shops and schools were closed. Day care facilities were only available for children of health care professionals and other essential professional groups. Further, various restrictions were imposed for many recreational activities, and socializing was limited, especially regarding social contact with people belonging to a risk group for severe COVID-19 outcomes. As a consequence, face-to-face contacts were widely replaced by digital social interactions (Suter, Külling, Zollinger, & Waller, 2021). The same was true for work-related interactions, as people were directed to work from home if possible. For parents, this often meant that paid work, childcare and home schooling of school-aged children had to be coordinated, both in terms of space and time. Overall, the pandemic rapidly changed the daily lives of children and their families in Switzerland and required a lot of flexibility from them, also because it was uncertain how the pandemic would unfold and how long it would last.
Today, a wide range of studies in different countries have indeed shown that these pandemic-related stressors had a negative impact on the populations’ well-being, mental health and behaviour (e.g. Organisation for Economic Co-Operation and Development [OECD], 2021; Orgilés, Morales, Delvecchio, Mazzeschi, & Espada, 2020; Ravens-Sieberer et al., 2022; Wissmath, Mast, Kraus, & Weibel, 2021). For example, stress and limited opportunities for social contact and activity, in addition to emotional and (psycho-)somatic complaints, can lead to more anger (Roy et al., 2020; Smith et al., 2021); in turn, the propensity for aggression can increase in adults as well as in children (Peterman et al., 2020). Regarding the pandemic’s impact on the well-being and behaviour of children, both distal (e.g. lockdown of social life) and more proximal factors (e.g. reduced face-to-face interaction with peers, grandparents, and teachers) come into play (Essler, Christner, & Paulus, 2021). For example, the closure of kindergartens and schools can lead to insecurity, social isolation, and a feeling of a lack of control in children, which can then lead them to experience frustration, anger, and despair (Stadler & Walitza, 2021). Consequently, they may show more oppositional behaviour and limit testing (Humphreys, Myint, & Zeanah, 2020). In addition, crowded living conditions can increase stress for children (Ravens-Sieberer et al., 2022). What impact these pandemic-related stressors actually had on children’s well-being and behaviour depends at least to some extent on the well-being of their parents or caregivers (e.g. Essler et al., 2021), who themselves were confronted with extraordinary challenges. In other words, during the COVID-19 pandemic, children were even more dependent on their caregivers’ ability to self-regulate and cope with the challenging situation. For example, many parents had to organise working remotely from home and taking care of their children at the same time (Fritschi & Fischer, 2020; Fuchs, Lanfranconi, Abbas, & Eckerlein, 2021). The resulting distress in child-parent interactions can lead to an additional increase in stress for children and parents alike, affecting their well-being (Essler et al., 2021).
Moreover, due to the more intensive intra-familial contact and the temporarily restricted direct extra-familial contact, there was a greater risk of family conflicts (Wako et al., 2015). Accordingly, in the spring of 2020, various parties already voiced a fear that family conflicts and violence would increase (e.g. Federal Coordinating Committee for Family Affairs [CCFA], 2020; World Health Organization [WHO], 2020); studies indicate that this indeed became true in many countries (e.g. Boserup, McKenney, & Elkbuli, 2020). In Switzerland there was a rise in family conflicts and violence towards children in certain phases of the pandemic (Bütikofer, Craviolini, Hermann, & Krähenbühl, 2020; Krüger & Caviezel Schmitz, 2020, 2022; Pro Juventute, 2021). This is particularly worrisome, as, at the same time, containment measures restricted access to important resources for children’s (emotional) well-being (e.g. leisure activities and social support by peers, grandparents, teachers, or others) (Mohler-Kuo, Dzemaili, Foster, Werlen, & Walitza, 2021; Stadler & Walitza, 2021).
Therefore, it can be assumed that the accumulation of constraints presented a challenge for many families. This heightened the risk of behavioural problems in children and parental strain, which in turn also affected their well-being in addition to the pandemic itself (Essler et al., 2021). Furthermore, from a developmental psychology perspective, younger children could have been particularly affected by the pandemic due to their higher dependency on their parents’ and caregivers’ care and support during challenging times (Essler et al., 2021; Marques de Miranda, Da Silva Athanasio, Sena Oliveira, & Simoes-E-Silva, 2020). In contrast to older children and adolescents, young children rely more on parents and caregivers for self-regulation and coping, as they have few(er) coping strategies due to their developmental stage. Accordingly, many studies in various countries revealed negative impacts of the pandemic on children’s psychological well-being, including high rates of depression and anxiety. Studies also found differences between age groups (for a summary, see Marques de Miranda et al., 2020). For example, whereas children aged 3 - 6 years were more likely to show clinginess and fear that family members might become infected with the Sars-CoV-2 virus, older children and adolescents more often showed inattention and persistent inquiry (Jiao et al., 2020, cited in Marques de Miranda et al., 2020).
The same is true for Switzerland (Lannen, Duss, Bombach, Graf, & Simoni, 2021; Mohler-Kuo et al., 2021; Schmidt, Barblan, Lory, & Landolt, 2021) even though containment measures were much more lenient than in other countries. For example, Schmidt et al. (2021) found that whereas adolescents showed the highest increase in emotional problems during the first lockdown in spring 2020, children aged 1 - 6 years showed more oppositional-defiant behaviours. In sum, it has been widely established that the pandemic had a negative impact especially on the well-being and mental health of children and adolescents and that these effects of the pandemic will have long-term consequences for the children’s development.
It is therefore particularly important to investigate the pandemic’s impact on children’s well-being and behaviour. However, until now, most studies have focused on children’s emotional well-being, depression, and/or anxiety. Relatively few studies have examined the pandemic’s impact on aggressive behaviour in children (Schmidt et al., 2021; Shorer & Leibovich, 2022; Sun et al., 2022). This study aimed to address this gap. Given the distal and proximal factors potentially increasing the risk for aggressive behaviour in children mentioned above, we aimed to examine, from the parents’ perspective, the impact of risk factors of aggressive behaviour in children heightened by the pandemic on the well-being, difficulties, and aggressive behaviour of children of different age groups. In detail, we hypothesised that parents’ burden of coordinating work and childcare responsibilities as well as the presence of a household member at higher risk for severe COVID-19 outcomes would have a negative impact on children’s well-being, difficulties, and aggressive behaviour. Further, we assumed that children living in single-parent households would have fewer opportunities for face-to-face interactions with different persons and would therefore be more affected by containment measures; they would be more likely to exhibit difficulties and show lower levels of well-being and more aggressive behaviour. In contrast, having access to a large balcony or garden/backyard should have had a positive impact on children’s well-being and behaviour.
The study presented here is part of the authors’ longitudinal study “Life in Times of Corona”, which examines short-, medium- and long-term effects of the pandemic on family conflict and violence in Switzerland using a nationwide online survey and secondary analyses of administrative data on family violence (e.g. Police Crime Statistics) (Krüger & Caviezel Schmitz, 2021, 2022). The sub-study “Children’s Life in Times of Corona” focused on the pandemic’s impact on children’s well-being, difficulties, and aggressive behaviour from the parents’ perspective.
2. Methods
2.1. Procedure and Instruments
On April 21, 2020, during the first nationwide lockdown in Switzerland, we launched an opt-in longitudinal online survey, asking parents of children aged 2 - 14 years about their children’s emotional well-being and behaviour during this extraordinary situation.1 We refrained from directly interviewing children and adolescents because young children would not have been able to fill in the questionnaire, and even for older children we would have needed the parents’ consent. Because the timing of the survey was critical, we chose to survey parents instead. We also ruled out surveying only adolescents aged 15 years and older, where parental consent would not be needed, because we expected the lockdown to have different impacts on younger and older children.
An online survey was created ad hoc and distributed via different paths: project web page, media reports, web pages and newsletters of relevant institutions and associations (such as educational counselling centres), and social media (Facebook, WhatsApp). We thus relied on a snowball sampling effect. In addition, members of the general population of German-speaking Switzerland who participated in the survey Life in Times of Corona and who lived (at least temporarily) in a household with underage children were also made aware of this online survey, Children’s Life in Times of Corona. The data presented here were collected from April 21 to June 7, 2020. This was the period of the first nationwide lockdown (March-April 2020) and closure of mandatory schools and childcare facilities (until May 11, 2020). Beginning in June, further easing of measures was decided, such as the re-opening of secondary and occupational schools, universities, and recreational facilities. However, a mask requirement on public transportation was introduced one month later. In this baseline survey, we asked parents for permission to contact them for a follow-up. If they agreed, they were contacted again at a later point in time. Results of this second wave in summer 2020 are presented elsewhere.
We collected data on socio-demographic characteristics of the children and their families, such as the children’s and respondents’ age, sex, and nationality. In addition, we included questions on the organisation of childcare and remote work from home during the lockdown. We also asked about the families’ housing situation, especially if they had access to a garden/backyard or larger balcony, and whether they themselves or another household member belonged to one of the COVID-19 higher risk groups.
To survey the children’s difficulties, (aggressive) behaviour and emotional well-being, we used mainly standardised scales and subscales, namely, the Inventory of Aggressive Behaviour in Children (FAVK-parent; Görtz-Dorten & Döpfner, 2010) to assess aggression triggering and sustaining factors in children aged 4 - 14 years and individual items and subscales of the German version of the Strengths and Difficulties Questionnaire (parent report) (SDQ; Goodman, 1999; Klasen, Woerner, Rothenberger, & Goodman, 2003) to assess emotional well-being, behaviour, and difficulties in children aged 2 - 14.
FAVK. The Inventory of Aggressive Behaviour in Children (FAVK; Görtz-Dorten & Döpfner, 2010) is a German questionnaire measuring four constructs that are assumed to play a role in children exhibiting peer- and adult-related aggressive symptoms, namely, social-cognitive information processing disorders (FAVK-social information processing [FAVK-SIP]; 16 items), impulse control disorders (FAVK-impulse; 12 items), disorders of social skills (FAVK-skills; 12 items), and disorders of social interactions (FAVK-interact; 10 items). In total, the questionnaire comprises 50 items. In the version used here (FAVK-parent), parents are asked to evaluate each of the items on a four-point Likert scale ranging from 0 (“not at all”) to 3 (“very much”) with respect to their child’s interactions with peers and adults. In this way scores on four different levels can be built with higher scores indicating greater dysfunctionality. In detail, scores can be built on an item group level for each construct separated for aggression towards peers and towards adults, and on the level of each of the four underlying constructs (subscale level). Finally, an aggregated aggression score can be built for aggression towards peers (FAVK-PEER; 25 items) as well as for aggression towards adults (FAVK-ADULT; 25 items). Based on these two aggression scores, an overall aggression score can be calculated. Cut-off values (on the item group level) and norm scores (for the four subscales, the aggregated aggression scores, and the overall aggression score) from a German norm sample of children aged 4 - 8 and 9 - 14 were available to interpret results (Görtz-Dorten & Döpfner, 2010). For our Swiss sample, we assumed that these norm values from Germany were applicable to children in German-speaking Switzerland due to the geographical and cultural proximity of both countries or regions. Because norm scores for the FAVK (Görtz-Dorten & Döpfner, 2010) were available only for children aged 4 - 14, only participants reporting at least one of their children as being that age were asked to complete the FAVK. In addition to using norm values, the distribution of achieved scores was compared with the distribution of scores in the norm sample in order to assess whether the proportions of children with clinically relevant scores in the sample corresponded to the expected proportions.
SDQ. In this study we used the German version of the SDQ (Klasen et al., 2003), a well-established and validated instrument to measure children’s strengths and difficulties. More precisely, we included the SDQ’s emotional symptoms subscale and the conduct problems subscale as well as the overall rating of whether the child had emotional or behavioural problems, when these problems started, and how much of a burden these difficulties were for the family (Woerner, Becker, & Rothenberger, 2004) . We asked participants for this overall evaluation for their children of all ages; it served as an estimate of current burden on the family. Participants with children aged 2 - 3 were asked to answer all items on the two SDQ subscales included, and parents of children aged 4 - 14 were asked to answer only the following two items from the emotional symptoms subscale in order to keep the response time reasonable: “Many worries, often seems worried”, “Often complains of headaches, stomach-aches or sickness”. Parents were asked to indicate their agreement with the items on a 3-point scale by marking the box for “not true”, “somewhat true”, or “certainly true”. The SDQ subscales were interpreted according to the instructions given by sdqinfo (2015). For the individual items on the SDQ and the overall evaluation of difficulties no norm values are available.
2.2. Data Analysis
Descriptive statistics were used to describe the families’ characteristics. Chi-squared tests were used to analyse differences in well-being, difficulties, and aggressive behaviour between children of different ages and sex as well as between children in families exhibiting different risk factors included in the study (Agresti & Finlay, 2009). However, as we surveyed a non-random sample, we could not statistically test our underlying hypotheses. However, reported p-values can be interpreted as an indication of where it might be worth taking a closer look in future studies using a random sample.
3. Results
3.1. Family Characteristics
Participants were biological parents, stepparents, or foster parents living at least part-time in a household with the child or children that they reported on. During the first lockdown until re-opening of compulsory schools and care facilities on May 11, 2020, 105 participants answered questions on the well-being and behaviour of 152 children aged 2 - 14. Another 54 participants reported on 79 children between May 11 and June 7, 2020, a period when compulsory schools were open again but other restrictions were still in place. Hence, a total of 159 participants reported on 231 children.
3.1.1. Socio-Demographic Data of the Respondents
The average age of the participating parents and caregivers, who (at least temporarily) lived with the child/ren they responded on, was 41.48 years (SD = 5.22; ranging from 32 - 64 years). The majority of respondents were women (83.6%), Swiss nationals (91.0%), and employed (93.5%), predominantly part-time (90.1%). Therefore, our results mostly reflected mothers’ perspectives on the children’s well-being and behaviour during the first period of the pandemic.
3.1.2. Organisation of Childcare and Remote Work from Home during the Lockdown and the Families’ Housing Situation
During the first lockdown, the majority of respondents cared for their children themself or together with their partner (91.0%). Less than 10% reported that their children were cared for by other family members or care facilities or that their children were old enough to look after themselves (7.7%). Two thirds of parents who have at least sometimes cared for their children themselves reported that they did this at least some of the time while working (65.5%). Not surprisingly, this was especially true for parents of older children. While about half of the parents with younger children at home (age 2 - 3: 50.0%; n = 9; age 4 - 8: 51.2%; n = 42) cared for their children at least partly while working, this was the case for more than three quarters of parents who (also) had older children to care for (age 9 - 14) (79.2%; n = 42).
The majority of respondents (86.5%) said they lived with their family (partner and children); only about 10% lived alone with their child/children. Accordingly, the proportion of single-parent families was around 10%. A total of 10.4% of respondents indicated that they themselves or one of their household members belonged to one of the known risk groups for severe outcomes of COVID-19. The majority of respondents (92.3%) stated that their home had a garden/backyard and/or a large balcony. Due to the lack of variations in these family characteristics, no differences in the children’s well-being, difficulties and behaviour were calculated between single-parent and non-single parent households, between families having a garden/backyard and/or a large balcony and those who did not have one, or between families with and without a family member at higher risk for severe COVID-19 outcomes.
3.1.3. Socio-Demographic Characteristics of the Children
The respondents’ children’s age ranged from 2 to 14 (M = 6.93 years, SD = 3.68 years, n = 231). The majority of children were between 4 and 8 years old (42.9%, n = 99), more than one third was between 9 and 14 years old (35.5%, n = 82) and one fifth of the children were 3 years old or younger (21.6%, n = 50). The sex ratio was balanced, with 111 boys and 120 girls. This was also true for the sex ratio within the different age categories (from 48.0% to 56.1% of children were girls).
To compare the different age groups on the perceived emotional well-being and behavioural problems, in the following we first summarise results for items or scales used for all age groups. We then report findings based on items and scales used for specific age groups.
3.2. Children’s Difficulties, Emotional Well-Being, and Somatic Symptoms (All Age Groups)
3.2.1. Children’s Difficulties
According to the respondents, 39.0% (n = 90) of the children showed emotional or behavioural problems. Most of these children (83.3%, n = 75) showed only mild difficulties, but 16.7% showed significant or major difficulties (n = 15). Interestingly, even though the majority reported only mild difficulties, the children’s emotional and behavioural problems were a burden for 91.1% (n = 82) of the families. In most cases (65.6%, n = 59) the burden was minor. Even though we cannot determine to what extent these problems were actually caused by the lockdown, it is noteworthy that in 19.1% (n = 17) of the cases these problems began during that time period. In the remaining cases they started one to six months prior to the survey (36.0%) or even more than six months prior (44.9%; n = 89).
Older children seemed to have difficulties more often than younger children (see Figure 1). Whereas the proportion of toddlers (aged 2 - 3) for whom difficulties were reported was 10.0%, difficulties were reported for 42.9% of children aged 4 - 8 and 52.4% of children aged 9 - 14 (χ2 (2, n = 230) = 24.483, p < 0.001). Boys were more likely to have difficulties than girls (45.0% vs. 33.6%; n = 230) (χ2 (1, n = 230) = 3.151, p = 0.076), and the proportion of children with significant difficulties was significantly greater for boys (10.8%) than for girls (2.5%) (n = 230) (χ2 (2, n = 230) = 7.458, p = 0.024). Accordingly, boys’ difficulties were perceived as a greater burden on the family than girls’ difficulties (assessed as at least a significant burden in terms of 34.0% of boys with difficulties vs. 15.0% of girls with difficulties, n = 90) (χ2 (1, n = 90) = 4.217, p = 0.040). In contrast, there were no significant differences between the sexes regarding how long the difficulties had existed (p = 0.916).
![]()
Figure 1. Overall difficulties separated by children’s age and sex (SDQ; n = 230).
There was a significant association between the perceived difficulties of the children and the organisation of childcare and remote work from home during the lockdown (χ2 (1, n = 208) = 8.615, p = 0.003). Respondents who cared for their children at least partly while working from home reported difficulties for 47.0% of children, whereas this was the case for only 26.3% of children whose parents cared for them without working at the same time.
3.2.2. Children’s Emotional Well-Being
Overall, a majority of children were perceived by their parents/caregivers as being able to cope well with the lockdown (see Figure 2). However, for overall 20.2% of the children, emotional problems (“Many worries, often seems worried”) during the lockdown were reported by their parents/caregivers. Children aged 9 and up seemed to have greater difficulties during the lockdown than younger children: For 37.2% of the children aged 9 - 14, at least some worry and distress were reported, whereas this was the case only for 13.7% of children aged 4 - 8 and for 6.0% of the toddlers (χ2 (2, n = 223) = 22.724, p < 0.001) (see Figure 2). There was no significant difference between the sexes (p = 0.505).
![]()
Figure 2. Children’s emotional well-being separated by children’s age and sex (SDQ; n = 223).
3.2.3. Children’s Somatic Symptoms
For somatic symptoms (“Often complains of headaches, stomach-aches or sickness”) the picture was the same as for emotional problems. Overall, 19.5% of the children displayed somatic symptoms during the lockdown. And again, children aged 9 and up seemed to exhibit more somatic symptoms than younger children (χ2 (2, n = 226) = 10.212, p = 0.006) (see Figure 3). According to their parents only 6.0% of toddlers complained of headaches, stomach-aches, or nausea at least some of the time, but the proportion was 18.8% for children aged 4 - 8 and 28.7% for children aged 9 - 14. Again, no significant difference between the sexes was found (p = 0.896).
3.3. Aggressive Behaviour in Children Aged 4 - 14
Regarding children aged 4 - 14, participants were asked to answer all items on the FAVK (Görtz-Dorten & Döpfner, 2010). The FAVK could be fully evaluated for 134 of the 181 children aged 4 - 14; these were 53.0% (n = 71) girls and 47.0% (n = 63) boys.
![]()
Figure 3. Children’s somatic symptoms separated by children’s age and sex (SDQ; n = 226).
Concerning the overall aggression score, clinically relevant scores were reported for 16.4% of the children. In comparison to the norm sample of children aged 4 - 14, children with a clinically relevant overall aggression score were overrepresented in our sample. Interestingly, 18.7% of the children showed clinically relevant scores for aggression towards peers (FAVK-PEER) and only 11.9% towards adults (FAVK-ADULT). Again, children with a clinically relevant aggression score were overrepresented in our sample.
The scores on the subscales revealed that the children in our sample had the most difficulties within the domains of social interactions (subscale FAVK-interact) and impulse control (subscale FAVK-impulse), even though none of the four subscales really stood out (see Figure 4). The proportion of children with clinically relevant scores in the sample was approximately within the expected range (10%) for the FAVK-SIP subscale (11.9%). For the FAVK-skills subscale, it was slightly above the expected proportion (14.9%), and for the FAVK-interact (17.9%) and the FAVK-impulse subscale (17.2%), the proportions of children with clinically relevant values in the sample were significantly above the expected proportion.
![]()
Figure 4. Scores on the FAVK subscales FAVK-SIP, FAVK-impulse, FAVK-skills and FAVK-interact separated by age group (n = 134 children aged 4 - 14).
On the item group level, the proportion of clinically relevant scores in the sample was 20% or more on the domain of impulse control towards adults (29.1%), on the domain of social skills towards peers (23.9%), and on the domain of social interactions towards peers (20.1%). Particularly noteworthy were the proportions of children with clinically relevant scores on the domain impulse control, where more children reached the cut-off for aggression towards adults (29.1%) compared to aggression towards peers (15.7%). This meant that parents and caregivers reported disturbances of impulse control in the behaviour of their children significantly more often towards adults than towards peers (see Table 1).
Age. In contrast to the results presented above for the SDQ measures, the FAVK resulted in a higher proportion of younger children with a clinically relevant overall aggression score (23.0% of the children aged 4 - 8 vs. 8.3% of the children aged 9 - 14) (χ2 (1, n = 134) = 5.175, p = 0.023). On the level of aggregated aggression scores towards peers and towards adults as well as on the subscale level, younger children were more affected: A higher proportion of children aged 4 - 8 showed a clinically relevant aggregated aggression score towards peers (28.4%) and towards adults (16.2%) than older children aged 9 - 14 (6.7% for aggression towards peers and for aggression towards adults, respectively) (FAVK-PEER: χ2 (1, n = 134) = 10.292, p = 0.001; FAVK-ADULT χ2 (1, n = 134) = 2.874, p = 0.090). However, it should be noted that the sex distribution across the age categories was not equal: Among the children aged 4 - 8, 48.6% (n = 36) were girls, whereas there were 58.3% (n = 35) girls among the children aged 9 - 14 (n = 134).
Sex. Twice as many boys (22.2%) as girls (11.3%) showed clinically relevant overall aggression scores. Sex differences were also evident on the other levels of analysis, with more boys (25.4%) than girls (12.7%), showing clinically relevant aggression towards peers scores as well as aggression towards adults scores (boys: 17.5%; girls: 7.0%; n = 134). The same was true for two of the four subscales,
![]()
Table 1. Results of the FAVK on the item group level (n = 134).
namely FAVK-skills and FAVK-interact (see Table 2). In sum, boys in our sample seemed more affected by aggression triggering and sustaining factors than girls. This was particularly true for disorders of social skills.
Overall, there was no significant difference between the sexes regarding the parents’ organisation of work and childcare (p = 0.563), with 62.3% (n = 43) of girls and 67.2% (n = 39) of boys being cared for by their parents at least sometimes while they were working (n = 127). In addition, there was no significant association between the parents’ organisation of work and childcare and children’s overall aggression scores (p = 0.442). However, a closer look at the children being cared for by their parents at least sometimes while working showed that more boys (25.6%; n = 10) than girls (11.6%; n = 5) had a clinically relevant overall aggression score, even though the difference was not statistically significant (p = 0.101). Furthermore, whereas 30.3% (n = 10) of children (aged 4 - 8) whose parents had to work at least partly while taking care of their children showed clinically relevant scores, this was true for only 10.2% (n = 5) of the older children (aged 9 - 14) (χ2 (1, n = 82) = 5.330, p = 0.021). However, when interpreting these differences between age groups and between sex groups, it is important to consider the unequal distribution of sex in the age categories.
3.4. Toddlers’ Emotional and Conduct Problems
Among toddlers in the sample, 8.3% (n = 4) achieved at least slightly raised scores in the area of emotional problems. Thus, compared to children of the same age in the norm sample (20%), toddlers with emotional problems were underrepresented in our sample. No differences between boys (8.0%) and girls (8.7%) could be found. Among toddlers in the sample, 20.8% (n = 10) achieved at least slightly raised scores for conduct problems. This corresponds with their proportion in the norm sample (sdqinfo, 2015). Again, no differences between the sexes could be found, with 22.7% of girls and 19.2% of boys aged 2 - 3 showing clinically relevant scores.
![]()
Table 2. Results of the FAVK on the sub-scale and aggregated scale level (n = 134).
4. Discussion
The aim of this study was to examine the emotional well-being, difficulties, and aggressive behaviour of children during the first phase of the COVID-19 pandemic in German-speaking Switzerland from the perspective of parents and caregivers, who lived (at least temporarily) together with the child/ren they reported on. In sum, our findings indicate that many children were able to cope well with the lockdown. However, for a significant number of children, parents reported difficulties, aggressive behaviour, and emotional or behavioural problems. About two-fifths of children showed difficulties. Emotional problems and somatic symptoms were reported for every fifth child. Even though the difficulties were mainly perceived as mild, they were a burden for the majority of families. In almost one fifth of the cases, the problems definitely began during the lockdown. Even though we have no information on the children’s well-being and behaviour prior to the pandemic, this is an important result, because there are indications that existing disadvantages could have been intensified by the pandemic, especially in families where parents were stressed and children showed behavioural problems (Langmeyer, Guglhör-Rudan, Naab, Urlen, & Winklhofer, 2020; Marques de Miranda et al., 2020). Moreover, children or families that were already disadvantaged could have been additionally harmed by the pandemic, because important support systems (family doctors, therapists, leisure activities, educational support, school social work, etc.) were no longer accessible or only accessible to a limited extent.
Compared to the norm sample (Görtz-Dorten & Döpfner, 2010), children with clinically relevant aggression scores were overrepresented in our study sample, with aggressive tendencies reported mainly towards peers and less towards adults. This is plausible, insofar as children probably try harder to avoid acting out towards their parents than towards their siblings or peers. At the same time, temper outbursts towards adults could have been particularly salient for the surveyed caregivers, which would be a possible explanation for why at the same time clinically relevant scores for impulse control disorders towards adults were reported particularly often. Overall, impulse control disorders and social interaction disorders were reported particularly frequently in our sample, which could be related to the fact that anomalies in these areas were easier to observe for parents and caregivers during the lockdown.
Even though the reported difficulties in children do not seem to be related to the caregivers’ employment per se, there was an association between these difficulties and the organisation of work and childcare responsibilities. For example, the proportion of children for whom difficulties were reported was almost twice as high in respondents who reported caring for their children at least partly while working than in respondents who reported caring for their children without working at the same time. This is in line with our hypothesis (see section 1) and could indicate that it is not so much the parents’ employment per se that plays a role regarding possible difficulties but rather the opportunity to adequately coordinate childcare and work. Working and (at least partially) caring for children at the same time can be a burden for parents, as various surveys in Switzerland (e.g. Bütikofer et al., 2020; Fritschi & Fischer, 2020; Lanfranconi, Gebhard, Lischer, & Safi, 2021) and other countries (Langmeyer et al., 2020) indicate. These challenges were especially due to school and childcare facility closures during the lockdown, as working parents had to reorganise their work, childcare, and their children’s schooling at short notice. For example, respondents in the study by Lannen et al. (2021) described managing remote work from home, home schooling older children, and caring for younger children at the same time as extremely challenging, with the needs of younger children sometimes having to be put aside in favour of home schooling older children. This stress on parents due to a wide range of demands may have led to distress in child-parent interactions and therefore also to stress on the children. In turn, this stress can lead to an increase in family and parental strain (Essler et al., 2021). This underlines how important it is for policy makers and also employers to take employees’ care responsibilities into account and to be more flexible, especially in times of pandemic when working remotely from home is mandatory.
However, to some extent our results—reflecting primarily mothers’ views on the children’s well-being and behaviour—might be explained by the fact that women were more affected than men by the pandemic independent of their educational level (Bütikofer et al., 2020; Kabeer, Razavi, & van der Meulen Rodgers, 2021).
Nevertheless, together with results from other studies, our findings indicate that already early on in the pandemic, containment measures had a direct effect on families’ situations and therefore an indirect effect on the well-being and behaviour of parents and children (Orgilés et al., 2020). However, the pandemic and related public health measures affected different children and families very differently, depending on the parents’ work situation or the children’s age and sex, for example. The latter is also in line with other studies (Schmidt et al., 2021), with older children showing the largest increase in emotional problems since the beginning of the COVID-19 pandemic and younger children showing the largest increase in oppositional-defiant behaviour. A plausible explanation for this would be that older children are more affected by lockdown restrictions, as the social radius of action usually increases with age. However, this picture was reversed for aggressive behaviour: Clinically relevant aggression scores were reported for a higher proportion of younger children (aged 4 - 8) than for older children (aged 9 - 14). A possible explanation could be that older children are more differentiated in their ability to report their feelings, needs, and difficulties to their parents or others than younger children are. The parents’ assessments of the children’s aggressive behaviour, on the other hand, are based more on observations and perceptions of the children’s behaviour. In addition, older children have better and more diverse coping and self-regulation skills, whereas younger children require more care and supervision, which means that parents may have had more opportunity to observe aggressive behaviour in younger children.
In addition to age, the children’s sex also seems to make a difference in the lockdown’s impact on their well-being and behaviour. Difficulties were reported more frequently and more strongly for boys than for girls. In addition, boys scored higher than girls on aggression triggering and sustaining factors. However, there were no differences between the sexes regarding the duration of the reported difficulties, suggesting that sex differences in distress were not related to lockdown. Instead, this could be an indication that the boys in our sample were already more stressed before the pandemic and that this vulnerability may have been increased by the lockdown. Furthermore, sex differences were found in relation to overall difficulties and aggressive tendencies but not for emotional problems or somatic symptoms, which may indicate that boys and girls were equally or differently burdened during the lockdown, depending on the area surveyed. However, it is also possible that results differ depending on how surveys are conducted. For example, in their sample of Swiss children and adolescents, Mohler-Kuo et al. (2021) found more ADHD-related symptoms and symptoms related to oppositional defiant disorder in girls than in boys. However, these differences in study results might be due to differences in the children’s self-perceptions and the parents’ perceptions of the children’s well-being and behaviour: Mohler-Kuo et al. (2021) surveyed children and adolescents, but we surveyed parents and other caregivers.
In sum, our findings indicate that even relatively short-term and more lenient public health measures can affect children negatively. This conclusion is in line with findings of other studies examining the impact of the lockdown on children (and their families) in Switzerland (e.g. Bringolf-Isler et al., 2021; Mohler-Kuo et al., 2021; Schmidt et al., 2021). How well children can cope with the challenges of lockdown and in the further course of the pandemic seems to depend on individual and family or parent-related resilience and risk factors (Shorer & Leibovich, 2022; Stadler & Walitza, 2021).
5. Limitations and Strengths
When interpreting the study results, some limitations have to be considered. First, we had a relatively small and non-random sample. In opt-in surveys, a self-selection bias is to be expected. As this is a non-random sample, the overrepresentation of children with clinically relevant scores could also be due to the fact that their parents or caregivers were more willing to take part in the survey. Second, the special features of the sample surveyed must be taken into account: The sample consisted mostly of mothers and employed persons. During the lockdown many participants had to coordinate working remotely from home and caring for their children. Therefore, our results mostly reflect mothers’ perspectives on the children’s well-being and behaviour. However, national and international studies indicate that working mothers were especially affected by the COVID-19 public health measures (Bütikofer et al., 2020; Kabeer et al., 2021; Lanfranconi et al., 2021) and the organisation of paid work, childcare, and home schooling was perceived as particularly challenging by many families (e.g. Fritschi & Fischer, 2020; Lannen et al., 2021). Third, due to the sample composition and the resulting lack of variation (e.g. regarding housing situation), correlations between children’s difficulties, aggressive behaviour, and emotional problems and possibly relevant variables could not be investigated. Further studies are needed.
Lastly, we relied solely on parental reports on difficulties, emotional and behavioural problems, and aggressive behaviour in their children. Thus, it is possible that our results are biased. Assessments could be influenced by the parents’ and caregivers’ own well-being, their situation, and the interactions with their children. For example, there might have been more opportunities to observe certain behaviour during the lockdown than before, or there might have been more tense situations at home due to handling work, childcare, and home schooling at the same time while having limited social interactions with persons outside the family at home. Nevertheless, our study results provide important insight into the parents’ perceptions of the well-being and behaviour of their children, and these perceptions affect the parents’ behaviour towards their children. What is perceived as a burden by individuals has real consequences and can, among other things, have an impact on the family climate and again therefore on children’s actual well-being (Essler et al., 2021; Orgilés et al., 2020; Shorer & Leibovich, 2022). This holds especially true for younger children, as they rely more on their caregivers’ behaviour as a basis of information for assessing unknown situations and also for coping with them than older children do. Furthermore, children strongly orient themselves to parents as models for emotion regulation, etc.
However, even though our study has some limitations, it also has some strengths. For one, the survey took place during the time of the lockdown restrictions. Thus, the data are not biased by memory effects. For another, this study is one of the few to focus on aggressive behaviour in children, and to our best knowledge, it is the only one using the FAVK.
6. Implications
Even relatively short-term and more lenient public health measures seem to affect children and their families. Behavioural difficulties and emotional problems represent a developmental risk for children as well as the family as a whole. Therefore, adequate prevention measures are needed. As children’s well-being and behaviour are closely linked to the well-being of their parents or families and vice versa, it is worth taking a holistic look not only at the needs and behaviour of children but also at the needs and well-being of caregivers and families. This is an important point, also regarding problematic parental behaviour such as child abuse and neglect or other types of family violence. And it is of special interest in times of pandemic, as long-lasting effects of the shelter in place orders on children, parents, and families must be taken into account. Consequently, when working with families, it is desirable to take such a holistic view and, if possible, to use evidence-based screening methods that make it possible to identify stress in the family system at an early stage.
Moreover, it must be considered that the pandemic and related public health measures affect children and families very differently—for example, depending on the children’s age or the parents’ work situation. This also shows that macrosocial structures play an important role in prevention. Accordingly, for example, policy makers are required to promote the recognition of childcare, and employers are responsible for creating family-friendly working conditions. Here, too, it must be considered that the needs can be very different and an intersectional perspective and (further) longitudinal data are needed to investigate the pandemic’s medium and long-term effects on children and families. Containment measures in the context of (future) pandemics ought to be increasingly assessed from the perspective of children and their families. The effects of containment measures may be eased, for example, by providing alternative options for recreational activities in public spaces, by expanding assistance provided by schools in the event of school closures and adapting it to the (heterogeneous) needs of families, and by offering low-threshold assistance and support services for families.
NOTES
1We also included parents of adolescents aged 15 - 17 years. However, data was reported for only 13 adolescents. Due to this small sample size, this age group was not included in our analysis presented here.